Urban Intersection Signal Control Method Based On Fluid Queuing Network

At present,the problem of urban road traffic congestion is frequent,which brings about time and energy waste,environmental pollution,and low travel efficiency.Signalized intersections are the key nodes of the urban transportation network.By properly configuring their signal control schemes,the traffic capacity of road sections,intersections and even the entire road network can be significantly improved,thereby effectively alleviating traffic congestion.Existing research on the optimization of intersection signal control schemes mostly focused on the analysis of the arrival characteristics of traffic flow,but lacked research on intersection service rules.How to accurately obtain performance metrics of the intersection to improve the effectiveness of signal control optimization is of great research value.In this paper,the fluid queuing model was used to accurately describe the entire process of traffic flow in all directions through the intersection.A queuing network model at the intersection was established,and a new adaptive signal timing optimization method was designed.The specific research work includes:（1）Establishment of the road section fluid queuing model.Considering the time-varying arrival rate of vehicles and the state-dependent service rate of system,the M/G（）/C/C queuing system of the section was established,and two models of vehicle speed were analyzed:linear model and exponential model.The M/G（）/1/C loss queuing model was established based on the fluid queuing theory.The impact of traffic jam propagation on the queuing system when encountering congestion was described,then the M/G（）/1/C feedback queuing model was established based on the loss queuing model and the system state equation was proposed.Finally,Simulink was used to build a discrete event simulation model to verify the effectiveness of the model under three scenarios of traffic demand.（2）Establishment of the intersection fluid queuing network model.In order to describe the right of way of vehicles at different signal phases,the service facilities at the intersection were divided into three types of nodes:imported road sections,entrance lanes and exported road sections according to the characteristics of traffic diversion and steering,and a dynamic topology of intersection queuing network was proposed.Then,with the system state equation discretized,several performance metrics of the intersection were derived through iteration:system state,output-flow rate,the probability of zero customers,and the customer blocking probability.Finally,Simulink was used to build an intersection discrete event simulation model to verify the effectiveness of the intersection queuing network model under two scenarios of traffic demand.（3）Establishment of the optimization model for intersection signal control and case analysis.Taking the average delay and average fuel consumption of vehicles as the optimization goals,and taking the green light duration,signal cycle and intersection saturation as constraints,an optimization model for the intersection signal control was established,and genetic algorithm was used to solve it.Finally,the intersection of Renmin West Road of Shudu Avenue-Dongchenggenshang Street was taken as an example to optimize the signal phase.The results showed that the optimized phase can reduce vehicle delay and fuel consumption to a certain extent,which proved the effectiveness of the optimization model for the intersection signal control.